Abstract About 65-80% of serious bacterial infections are biofilm-mediated. Not only do biofilms provide an anchor and physical protection from the immune system for bacterial cells, but the physiology and genetic programming of bacteria also shifts between the planktonic (free floating) and sessile (stationary) states. Most notably, antibiotic sensitivity differs substantially between the two states, with bacteria being orders of magnitude less sensitive to antibiotics in the sessile state. Biofilms are known to include a variety of polymers and proteins. One of these proteins has previously been shown to anchor the three dimensional scaffolding of the polymers. Trellis has used its proprietary antibody discovery technology to clone a high affinity antibody from human B lymphocytes, TRL1068, which binds the homologs of this protein from both Gram positive and Gram negative bacteria. In Phase I of this project, we used two rodent models of bacterial infection to demonstrate that extraction of the protein from the biofilm by this antibody leads to the biofilm dissolving in vivo as it does in vitro. Therapeutic benefit was seen for treatment of MRSA in both models: a rat model of infective endocarditis and a mouse model of implant infection. In Phase II, we will conduct IND-enabling studies for FDA approval by establishing a manufacturing method under GMP and conducting toxicity testing of the therapeutic antibody. In parallel, we will continue preclinical research to explore alternative indications and dosing regimens. This antibody offers potential clinical benefit against a wide range of infections that are currently very difficult to treat. Infective endocarditis in particular is a an indication for which current therapy often fails, leading to expensive heart valve replacement surgery that has a significant relapse rate (re-establishment of the biofilm protected infection) leading to death. Because the biofilm can be imaged directly (using trans-esophageal echocardiogram technology), efficacy defined by clinical endpoints can be correlated with the mechanism of action. Since the heart valve is readily accessible to antibody delivered intravenously, tissue penetration is not a significant variable for this indication. The combination of high unmet medical need and favorable experimental features makes this indication particularly useful for our initial clinical trials.